Journal of Cleaner Production, Volumen 174, 2018, Pages 807-820
Identifier
dc.identifier.issn
09596526
Identifier
dc.identifier.other
10.1016/j.jclepro.2017.11.025
Identifier
dc.identifier.uri
https://repositorio.uchile.cl/handle/2250/169355
Abstract
dc.description.abstract
An Eco-Industrial Park (EIP) is a community of businesses that seeks to reduce the global impact by
sharing material. The connections among the industrial participants within this park improve the
environmental performance of the industrial network. However, the connectivity also propagates failures. This risk is an important point of criticism and a barrier to industrial plants when evaluate their
integration to an EIP. This paper proposes an indicator to follow the resilience of an EIP so as to improve
the security of the whole system, considering the dynamic of the participants to endure a disruptive
event. This metric could be used by decision-makers in order to include the resilience in the design phase
of an EIP. Solving these security problems would expand the set of experiences of cleaner production,
facilitating the integration of industrial processes. The proposed resilience indicator is based on two main
characteristics of an industrial network: the number of connections among participants, and the capacity
of each flow to change its magnitude when a participant suddenly stops sharing flows within the park. A
network is separated in independent layers to quantify its flexibility when substituting flows. Each layer
includes a single shared material. The resilience of a multi-layer park is then calculated as a weighted
summation. This indicator is applied over two illustrative cases to study: Kalundborg, in Denmark; and
Ulsan, in South Korea. These applications show consistent results when compared with reality. Although
the proposed resilience indicator has been developed for material networks, it can be adapted to heat
integration networks. In this case, special attention should be payed to physical constraints as minimal
temperature gradients.